Forage Estrogens, Relative Potencies of Several Estrogenlike

Food Chem. , 1962, 10 (5), pp 410–412. DOI: 10.1021/jf60123a018. Publication Date: May 1962. ACS Legacy Archive. Cite this:J. Agric. Food Chem. 10, ...
0 downloads 0 Views 366KB Size
and Lomb Spectronic 20 colorimeter is shown in Figure 2. T h e slope of the graph is -0.1323.

Sample Analysis. Weigh into a 250ml. beaker 25.0 grams of milk or other dairy product, diluted until its protein concentration is equivalent to that of milk (about 37,). Add 5 ml. of 47, Rivanol and allow the mixture to stand 5 minutes for maximum protein precipitation. Remove the aggregated protein by filtration through folded filter paper. T o the filtrate add 1 drop of 10% sodium hydroxide and proceed as directed under “Preparation of Calibration Curve.” Determine the concentration of lactose from the calibration curve. Testing of Method. T h e method was tested for accuracy and precision by increasing the lactose concentration of a skim milk sample by from 1 to 6 7 6 with added lactose. T h e concentration of lactose was then determined by the described procedure. A single determination was made on each concentration shown in Table 11. O n the basis of these data the accuracy of the method in the concentration range investigated is within 10.27,. T h e largest deviation was 4.7y0 greater than the true value. Effect of Sucrose. In some dairy products sucrose is used as a sweetening agent. T o determine the effect of added sucrose on the measurement of lactose, 5 , 10, 15, and 20% commercial sucrose was added to a skim milk sample and the lactose determined. Allowing for the dilution due to the added sucrose, the amount of lactose found and the amount calculated to be present a t each concentration of sucrose are shown in Table 111.

The amounts of lactose determined are in good agreement with the calculated amounts. Therefore: the method can be applied to the analyses of dairy products containing sucrose.

Discussion T h e linear relationship between the amount of unreacted Fehling solution and the concentration of lactose points to the existence of a similar relationship between the amount of reacted Fehling solution and the concentration of lactose. By the hlunson-IYalker method. the amount of precipitated cuprous oxide retained on the asbestos mat is used as a measure of the extent of the reaction. T h a t the amount of cuprous oxide retained on the filter from the hot reacting solution is not stoichiometrically related to sugar content was observed in these analyses by the appearance of precipitated cuprous oxide in the cold filtrate. This filtrate has been in contact with borate ion and the reducing action stopped, as found by the measurement of identical absorbance of samples of filtrate taken before and after the appearance of precipitated cuprous oxide. O n the basis of these observations, cuprous oxide in a strong alkaline medium in the presence of tartrate has a definite solubility when hot. This fact may explain the need for such careful reproduction of all operations in successfully using the Munson-Walker procedure. Procedural difficulties which gave technicians the most trouble were in the removal of excess Rivanol after precipitation of the proteins. T h e amount of

Rivanol remaining depends upon the protein content of the product being examined. This Rivanol is precipitated by the addition of a small amount of base. If too much base is added. a cloudy suspension is produced which is not removed by filtration and causes low results. Its nature suggests that it contains some proteins. I t is obtained only when dealing Lvith dairy products. If the p H is raised too far, the precipitate reverts to a suspension which reacts with the cupric ion of Fehling solution. The usual appearance of a slight yello\v color does not affect the resulrs. as long as the filtrate is clear of suspended material. This method has the same limitations as the Munson-\Valker method. in that it is not specific for lactose. .411 reducing sugars present in the material analyzed \vi11 react with the reagent used.

Literature Cited (1) Association of Official Agricultural Chemists, “Official Methods of Analysis,” 8th ed., 1955. (2) Jenness, R., Patton, S., “Principles of Dairy Chemistry,“ pp. 120-2, Wiley, New York, 1959. (3) Kenyon, A. J.. Anderson, R. K., Jenness. R., J . Datrl Sci. 42, 1233 (1959). (4) Pigman. \V. W., Goepp, R. M . , Jr., “Chemistry of the Carbohydrates.” pp. 183-4. Academic Press, New York, 1748. Receivedfor review September 5, 1961. Accepted December 7. 1961. The use of trade names is f o r the purpose of identi$cation only and does not imply endorsement of the product or its manufnrt u w b.y the c‘.S. Department of Agriculture.

FORAGE EST R 0 0 E N S

Relative Potencies of Several EstrogenLike Compounds Found in Forages

A

RED CLOVER, and subterranean clover have been shown to contain a t least five compounds (2. 3, 73. 75) which can produce a n estrogenlike response in the mouse-coumestrol. genistein, biochanin .4, formononetin. and daidzein. All, with the exception of biochanin A, have also been found in ladino clover. Each compound has been evaluated a t various times by different workers (5-70, 77, 78). H o ~ v ever, since the bioassay techniques varied considerably between laboratories, it may be difficult or impossible to correlate results. LFALFA,

410

Bioassays of plants normally measure only total activity. T o provide data which will enable us to determine the relative contribution of each compound to the total uterine stimulating activity of a forage, it is necessary to know the relative amounts of each present as well as their relative potencies. T h e objective of the research reported herein was to provide data on the relative potencies of each of the five estrogen-like compounds, employing diethylstilbestrol and estrone as standards.

AGRICULTURAL AND F O O D CHEMISTRY

E. M. BICKOFF, A. 1. LIVINGSTON, A. P. HENDRICKSON, and A. N. BOOTH Western Regional Research laboratory, Albany, Calif.

Experimental Preparation of Compounds. T h e coumestrol and coumestrol acetate were prepared synthetically by a previously published method (72). Formononetin was isolated from red clover, and its identity confirmed by comparison of its physical properties with those of an authentic sample. Daidzein was ohtained from formononetin by demethylation. Genistein (m.p. 301’ C. decomp.) was isolated from subterranean clover. Biochanin A was synthesized according

The relative estrogenic potencies of five compounds found in forages-coumestrol and its diacetate, genistein, biochanin A, formononetin, and daidzein-have been evaluated b y a standardized mouse uterine weight bioassay procedure. For comparative purposes, diethylstilbestrol and estrone were also evaluated. Coumestrol was found to be 35 times as potent as genistein. Genistein was slightly more potent than daidzein, about twice as potent as biochanin A, and four times as potent as formononetin. to the procedure of Yoder et ai. (79). Diethylstilbestrol and estrone \\.ere commercially obtained materials. Bioassay. Immature female mice, 19 to 21 days old and weighing between 8 and 10 grams, were obtained commercially in lots of 50. The groups of mice bvere! for the rnost part, randomly divided into lots of 5. In a few cases: an extra mouse or two was obtained and added to the experiment. In some cases, a mouse may have been lost due to accident or death, which would also lead to deviation from multiples of 5. The compounds under test were dissolved in acrtone and added to the control diet; the solvent was removed under a current of air with uniform mixing. The details of bioassay were then identical with those previously described for forage extracts (7). Briefly, the test consists of feeding the mice for 4 to 6 days until an average of 10 grams of diet is consumed per mouse. The mice are then sacrificed, and the intact uteri are excised and weighed Lvithout blotting. T h e figures reported here represent pooled data and were obtained from a series of assays extending over a 2-year period. The results were examined for time effects and, if such existed. they were not great enough to affect interpretation of the results. Thc standard errors shown apply to pooled data and were calculated separately for each dose level. Dose-Response Relationship. The initial plan \vas to evaluate the relative potencies of the several compounds by means of slope-ratio assays? since for coumestrol. estrone, and diethylstilbestrol, the logarithms of the uterine weights were linear with dosage. However? for the isoflavones this was not true. Instead, a linear relationship existed between the mean uterine weights and the dosage. Therefore, an alternate procedure was chosen for comparing potencies. For each compound, dose response curves were plotted from the data in column 4 of Table I. The respective doses for each compound required to produce a uterine \\.eight of 25 mg. were found from these curves and are presented in column 5 of Table I . Relative potencies. defined as the ratio of the doses shown in column 5. to diethylstilbestrol, which is arbitrarily assigned a value of 100,000, are tabulated for each compound in column 6 of Table I . Slightly different estimates of relative potencies would be obtained if one picked a different response point.

Table 1.

Dose-Response Data and Relative Potency of Forage Estrogens vs. Diethylstilbestrol and Estrone

Compound

Quontitv Fed pe; Mouse, pg.

Number Uterine Weight Quonfitv to of iS.E.,“ Produce 25 Mice M g . Uterus, pg. Mg.

Relative Potencyb

Control Diethylstilbestrol

0.000 300 16 0.025 0.050 50 40 0.075 0.100 491 0.200 31

9 . 6 f0.3 11.8 f 0 2 18.1 f 0.7 23.7 f 0 8 29 2 f 0 . 4 78.1 f 6 2

Estronr

0.50

23 15 22 40 5

14.7 f 0 . 9 16.0 & 1 . 3 23.8 f 2 . 5 36.1 f 2 3 45.3 i 6 7 13.8 f 0 . 2 24.2 f 0 . 2 29.2 i 1 . 3 40.7 f 1 . 9 76.0 f 6 . 0

240

35

19.0 i 1.2

340

24

0.75

1 .oo

1 .so 2.00 Coumrstrol

100 200 300 400 500

20 44 126 79 36

Coumestroldiacetate

250 300 400 500

30 60 65 60

1 >ooo Genistrin

5,000

7,500 8.000 12.000 15.000 20,000 Daidzein

5,000 7,500 10,000

15,000 Biochanin A

Formononetin

10,000 20,000 30,000 40.000 15.000 20.000 25,000 30.000 40.000

0.083

100,000

1.20

6 900 ~

87

22.3 26.7 42.4 87.0

18 15 19 25 35 25

19.4 f 0 . 9 28.0 f 5 . 0 27 0 i 1 . 6 32.4 i 1 . 9 36.6 i 0 6 52.7 f 1 9

8,000

1 .oo

20 15 20 33

17.3 & 1 . 3 18.5 f 2 . 6 24.8 i 1 . 3 31.2 i 1 4

11.000

0.75

30 59

20 3 i 2 . 5 27 9 f 2 . 1 27.9 i 2 . 6 45 5 + 3 . 8

18.000

0.46

1 6 . 8 =t0 . 8 17.9 f 1 . 4 23.2 f 1 . O 27.5 f 2 . 3 26 1 i 2 . 0

32,000

0.26

20

38 20 15 19

20 15

f 1.6

f1.4 & 3.3 f 6.2

S.E.-Standard Error mean. Measured at the dosage required to produce a 25-mg. uterus.

Results

Genistein was the most potent of the isoflavones, while formononetin was the least potent (Table I). Previous studies at this laboratory (3) have shown that methylation of coumestrol reduces its estrogen-like potency. This is confirmed again in this study n i t h the isoflavones-i.e., biochanin A (4’methyl ether of genistein) was about one-half as potent as genistein. Lvhile formononetin (4’-methvl ether of daidzein) was one-third as potent as daidzein. The difference bet\\;ef=n the potencies VOL.

of genistein and daidzein is apparently d u e to the presence of the 5:hydroxyl group of genistein. Bradbury and White ( 8 ) suggested that the presence of the hydroxyl group in the 5- position modifies the properties of the carbonyl group. This type of effect: previously ascribed to hydrogen bonding and chelate-ring formation, has been shown by Hergert and Kurth (73) to be due partly to mesomerism which decreases the doublebond character of the carbonyl >C-0 /

linkage by partial

i o , NO. 5 ,

SEPT.-OCT.

Of

1962

“-O //

411

The diacetate of coumestrol was about equal in activity to coumestrol, taking into account the weight of the two acetate groups. Cheng and coworkers ( 7 7 ) under their bioassay conditions found daidzein to be the most potent of the four isoflavones, while biochanin A and genistein had about equal potencies. This apparent variation in results between that of Cheng’s laboratory and that included in this study may be due to differences in bioassay procedure. One major difference between their procedure and ours is that in their procedure the excised mouse uteri were fixed in Bouin’s fluid. then blotted before weighing. The authors found formononetin to have definite activitv, although less than that of other isoflavones. This is contrary to the findings of a t least three other laboratories (5, 8, 78) which reported formononetin to be inactive by either oral administration or subcutaneous injection a t levels of 10 to 50 mg. per mouse. Cheng ef ai. ( 7 7 ) reported that the feeding of formononetin to mice a t a level of 2.5 mg. per day for 4 days produced a slight increase in uterine weight. Our results, Table I, show that coumestrol is from 30 to 100 times more potent than the isoflavones. Estrone, the natural animal estrogen, and the synthetic diethylstilbestrol are respectively about 200 and 3000 times more potent than coumestrol under the conditions of our bioassay. Despite the lower potencier of the Tor age estrogens, when compared to diethylstilbestrol. they are present in

certain forages in sufficient quantities as to be influential in animal nutrition and physiology. Thus, Oldfield et al. (76) have shown that alfalfa meal, selected for high coumestrol content, can cause increased weight gains in wether lambs. Furthermore, the even less potent isoflavones, notably genistein, have been shown to be present in Australian subterranean clover ( 7 ) in sufficient quantities to cause reproductive problems in sheep. Acknowledgment

The authors are indebted to E. J. Cnderivood, Director, Institute of Agriculture. University of Western Australia, Sedlands. Australia, foi the genistein employed in this investigation; to E. D. \$’alter of this laboratory for preparation of formononetin from red clover; to E. M’. Cheng (deceased), foimerly of Iowa State University, for authentic samples of each of the four isoflavones to serve as reference compounds; and to Ann Gramps for preparing samples for assay. literature Cited

(1) Bickoff, E. M., Booth, A. N., Livingston, A4.L.? Hendrickson, A. P., Lyman, R. L.? J . Animal Sci. 18, 1000 (1 959). (2) Bickoff, E . M.: Booth, A. N., Lyman, R. L.? Livingston, A. L.? Thompson, C. R., DeEds, F., Science 126, 969 (1 957). (3) Bickoff, E. $1.. Booth. A . S . , Lyman, R. L., Livingston, ,4. L., Thompson, C. R.. Kohler, G. O.,

J. AGR.FOODCHEM.6, 536 (1958). (4) Bickoff, E. M., Livingston, ‘4. L.? Booth, .A. N., Arch. Biochem. Biophys. 88, 262 (1960). (5) Biggers, J. D., Curnow, D. H., Biochem. J . 58, 278 (1954). (6) Eose, J. L., Chandran, K.. J . Sei. Znd. Research (India) 13B, 888 (1954). ( 7 ) Bradbury, R. B.: White, D. E., J . Chem. SOL.1951, 3447. (8) Bradbury, R. B., White, D . E., Vitamins and Hormones 12, 207 (1954). (9) Carter, M. W., Smart, Mi. W. G., Matrone, C., R o c . SOC.Expti. Bioi. Med. 84, 506 (1953). (10) Cheng, E., Story, C. D., Yoder, L., Hale, W. H., Burroughs, W., Science 118, 164 (1953). (11) Cheng, E., Yoder, L., Story, C. D., Zbid.,120, 575 (1954). Burroughs. W.: (12) Emerson, 0. H., Bickoff, E. M., J . A m . Chem. So:. 80, 4381 (1958). (13) Guggolz. J., Livingston. A . L., Bickoff, E. M., J. AGR. FOOD CHEY. 9, 330 (1961). (14) Hergert, H. L., Kurth, E. F.: J . Am. Chcm. Soc. 75, 1622 (1953). (15) Lyman, R. L., Bickoff, E. M., Booth, A . N., Livingston, A. L.: Arch. Biochem. Biophys. 80, 61 (1959). (16) Oldfield, J. E., Fox, C. W., Bickoff, E. hl., J . Animal Sei. 19, 1281 (1960). (17) Pope, G. S., Dairy Sci. Abstr. 16, 334 (i954). 118) PoDe. G. S.. Elcoate. P. V.. Simoson, i. ’.4., Andrew, D . G.; Cheh. Znd. (London) 1953, 1092. (19) Yoder, L., Cheng, E. Burroughs, \V.: Proc. I G W Z A C U ~ 61, . S C271 ~ . (1954). \

,

Received f o r reciew June 26, 1967. -4ccepted October 27, 7961. The ?Vestern Research Laboratory is a laboratory of the Western C’tiliratioti Research and Deuelopment Dir’ision, Agricultural Rtsearch Seruice, C:. S. Department of Agriculture.

COTTONSEED P R O T E I N S

Isolation and Chromatographic Characterization of Low Molecular Weight Cottonseed Proteins

C

PROTEIN is the term assigned to the composite mixture of individual proteins found in the defatted seed. Its value for food or feed is dependent upon the chemical content and physical characteristics of the individual proteins. Jones and Csonka ( 6 ) , using classical fractionation procedures, isolated six fractions from cottonseed. The a- and /3-globulins, glutelin, and pentose protein composed the major fractions. Protamine-type protein was absent. OTTONEED

412

Spies and coworkers (72: 73) have found the relatively minor water-soluble fraction to consist of a series of proteins varying widely in pentose content. Since lysine is a major determinant of the nutritive value of cottonseed protein ( 9 ) , the isolation of a lysine-rich fraction in any comprehensive study would be of particular interest. Both the pentose protein of Jones and the water-soluble fraction of Spies were seemingly high in the basic amino acids. LVith this carbohydrate content and

AGRICULTURAL AND FOOD CHEMISTRY

WILDA H. MARTINEZ and VERNON L. FRAMPTON Southern Regional Research Laboratory, Southern Utilization Research and Development Division, U. S. Department of Agriculture, N e w Orleans, La.

salt solubility of cottonseed protein in mind, a preliminary fractionation attempt was made using a SaC1-aqueous ethanol solvent system originally designed to dissociate and solubilize nucleoproteins (2). This resulted in the isolation of a fraction which was 1570 higher than the average protein in lysine content. This investigation. therefore, is concerned with the isolation and characterization of that portion of glandless cottonseed ( 8 ) \vhich is soluble in high molarity